{"id":13858,"date":"2015-04-14T23:22:16","date_gmt":"2015-04-14T20:22:16","guid":{"rendered":"http:\/\/hgpu.org\/?p=13858"},"modified":"2015-04-14T23:22:16","modified_gmt":"2015-04-14T20:22:16","slug":"image-classification-with-pyramid-representation-and-rotated-data-augmentation-on-torch-7","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=13858","title":{"rendered":"Image Classification with Pyramid Representation and Rotated Data Augmentation on Torch 7"},"content":{"rendered":"<p>This project classifies images in Tiny ImageNet Challenge, a dataset with 200 classes and 500 training examples for each class. Three network architectures are experimented: a traditional architecture with 4 convolutional layers + 2 fully-connected layers; a Tiny GoogleNet with 3 inception layers; and a pyramid representation-based network. Tiny GoogleNet achieved the highest top-1 validation accuracy of 47%. Work is done to reduce overfitting. Dropout improves validation accuracy by 10%. Data-augmentation of random crop and horizontal flip increased validation accuracy by 10%. Rotation does not appear to improve validation accuracy. Pyramid representation shows significant computational efficiency, achieving similar top result 240% faster computation time per batch. Training accuracy converges at 65 &#8211; 70% for all three networks. Future work is to increase expressive power of network. Training was done on Torch 7 with Facebook&#8217;s Deep Learning Extension.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>This project classifies images in Tiny ImageNet Challenge, a dataset with 200 classes and 500 training examples for each class. Three network architectures are experimented: a traditional architecture with 4 convolutional layers + 2 fully-connected layers; a Tiny GoogleNet with 3 inception layers; and a pyramid representation-based network. Tiny GoogleNet achieved the highest top-1 validation [&hellip;]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[11,89,3],"tags":[1782,14,1673,1025,34,20,1732],"class_list":["post-13858","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-nvidia-cuda","category-paper","tag-computer-science","tag-cuda","tag-deep-learning","tag-machine-learning","tag-neural-networks","tag-nvidia","tag-nvidia-grid-k520"],"views":3367,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/13858","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=13858"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/13858\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=13858"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=13858"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=13858"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}